Spaces:
Paused
Paused
Update app.py
Browse files
app.py
CHANGED
@@ -92,19 +92,33 @@ def upload_imgorvideo(gr_video, gr_img, text_input, chat_state,chatbot,audio_fla
|
|
92 |
elif gr_img is not None and gr_video is None:
|
93 |
print(gr_img)
|
94 |
chatbot = chatbot + [((gr_img,), None)]
|
95 |
-
chat_state
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
img_list = []
|
97 |
llm_message = chat.upload_img(gr_img, chat_state, img_list)
|
98 |
return gr.update(interactive=False), gr.update(interactive=False), gr.update(interactive=True, placeholder='Type and press Enter'), gr.update(value="Start Chatting", interactive=False), chat_state, img_list,chatbot
|
99 |
elif gr_video is not None and gr_img is None:
|
100 |
print(gr_video)
|
101 |
chatbot = chatbot + [((gr_video,), None)]
|
102 |
-
chat_state
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
103 |
img_list = []
|
104 |
-
|
105 |
-
llm_message = chat.upload_video(gr_video, chat_state, img_list)
|
106 |
-
else:
|
107 |
-
llm_message = chat.upload_video_without_audio(gr_video, chat_state, img_list)
|
108 |
return gr.update(interactive=False), gr.update(interactive=False), gr.update(interactive=True, placeholder='Type and press Enter'), gr.update(value="Start Chatting", interactive=False), chat_state, img_list,chatbot
|
109 |
else:
|
110 |
# img_list = []
|
@@ -121,15 +135,16 @@ def gradio_ask(user_message, chatbot, chat_state):
|
|
121 |
def gradio_answer(chatbot, chat_state, img_list, num_beams, temperature):
|
122 |
llm_message = chat.answer(conv=chat_state,
|
123 |
img_list=img_list,
|
124 |
-
num_beams=
|
125 |
temperature=temperature,
|
126 |
-
max_new_tokens=
|
127 |
-
max_length=
|
128 |
chatbot[-1][1] = llm_message
|
129 |
print(chat_state.get_prompt())
|
130 |
print(chat_state)
|
131 |
return chatbot, chat_state, img_list
|
132 |
|
|
|
133 |
title = """
|
134 |
<h1 align="center"><a href="https://github.com/DAMO-NLP-SG/Video-LLaMA"><img src="https://s1.ax1x.com/2023/05/22/p9oQ0FP.jpg", alt="Video-LLaMA" border="0" style="margin: 0 auto; height: 200px;" /></a> </h1>
|
135 |
|
|
|
92 |
elif gr_img is not None and gr_video is None:
|
93 |
print(gr_img)
|
94 |
chatbot = chatbot + [((gr_img,), None)]
|
95 |
+
chat_state = Conversation(
|
96 |
+
system= "You are able to understand the visual content that the user provides."
|
97 |
+
"Follow the instructions carefully and explain your answers in detail.",
|
98 |
+
roles=("Human", "Assistant"),
|
99 |
+
messages=[],
|
100 |
+
offset=0,
|
101 |
+
sep_style=SeparatorStyle.SINGLE,
|
102 |
+
sep="###",
|
103 |
+
)
|
104 |
img_list = []
|
105 |
llm_message = chat.upload_img(gr_img, chat_state, img_list)
|
106 |
return gr.update(interactive=False), gr.update(interactive=False), gr.update(interactive=True, placeholder='Type and press Enter'), gr.update(value="Start Chatting", interactive=False), chat_state, img_list,chatbot
|
107 |
elif gr_video is not None and gr_img is None:
|
108 |
print(gr_video)
|
109 |
chatbot = chatbot + [((gr_video,), None)]
|
110 |
+
chat_state = default_conversation.copy()
|
111 |
+
chat_state = Conversation(
|
112 |
+
system= "You are able to understand the visual content that the user provides."
|
113 |
+
"Follow the instructions carefully and explain your answers in detail.",
|
114 |
+
roles=("Human", "Assistant"),
|
115 |
+
messages=[],
|
116 |
+
offset=0,
|
117 |
+
sep_style=SeparatorStyle.SINGLE,
|
118 |
+
sep="###",
|
119 |
+
)
|
120 |
img_list = []
|
121 |
+
llm_message = chat.upload_video(gr_video, chat_state, img_list)
|
|
|
|
|
|
|
122 |
return gr.update(interactive=False), gr.update(interactive=False), gr.update(interactive=True, placeholder='Type and press Enter'), gr.update(value="Start Chatting", interactive=False), chat_state, img_list,chatbot
|
123 |
else:
|
124 |
# img_list = []
|
|
|
135 |
def gradio_answer(chatbot, chat_state, img_list, num_beams, temperature):
|
136 |
llm_message = chat.answer(conv=chat_state,
|
137 |
img_list=img_list,
|
138 |
+
num_beams=1,
|
139 |
temperature=temperature,
|
140 |
+
max_new_tokens=240,
|
141 |
+
max_length=511)[0]
|
142 |
chatbot[-1][1] = llm_message
|
143 |
print(chat_state.get_prompt())
|
144 |
print(chat_state)
|
145 |
return chatbot, chat_state, img_list
|
146 |
|
147 |
+
|
148 |
title = """
|
149 |
<h1 align="center"><a href="https://github.com/DAMO-NLP-SG/Video-LLaMA"><img src="https://s1.ax1x.com/2023/05/22/p9oQ0FP.jpg", alt="Video-LLaMA" border="0" style="margin: 0 auto; height: 200px;" /></a> </h1>
|
150 |
|